GSE215252 Processing Pipeline

Publication

BMC bioinformatics (2023) — PMID 37784060

Dataset

FLARE: A fast and flexible peak-calling pipeline for identifying RNA editing foci

1. 1

   STAR aligned original fastqs to hg19 reference build

   STAR v2.7.10a

   $ Bash example

   # Install STAR (example using conda)
   # conda create -n star_env star=2.7.10a -c bioconda -c conda-forge
   # conda activate star_env
   
   # --- Reference Genome Preparation (if not already done) ---
   # This step generates the STAR genome index for hg19.
   # Replace /path/to/hg19.fa and /path/to/hg19.gtf with actual paths.
   # The --sjdbOverhang parameter should be set to (ReadLength - 1) or 100 for typical RNA-seq.
   # For eCLIP, where reads are often short and unspliced, --sjdbOverhang might be less critical or set to a small value.
   # GENOME_DIR="/path/to/STAR_index/hg19"
   # STAR --runThreadN 8 --runMode genomeGenerate \
   #      --genomeDir ${GENOME_DIR} \
   #      --genomeFastaFiles /path/to/hg19.fa \
   #      --sjdbGTFfile /path/to/hg19.gtf \
   #      --sjdbOverhang 100
   
   # --- STAR Alignment Command ---
   # Define variables
   # Replace with actual paths and filenames
   GENOME_DIR="/path/to/STAR_index/hg19" # Path to the pre-built STAR genome index for hg19
   READ1="sample_R1.fastq.gz" # Input FASTQ file for Read 1
   READ2="sample_R2.fastq.gz" # Input FASTQ file for Read 2 (remove if single-end)
   OUTPUT_PREFIX="sample_aligned" # Prefix for output files
   THREADS=8 # Number of threads to use
   
   # STAR alignment command (parameters are common for eCLIP-like assays)
   STAR --runThreadN ${THREADS} \
        --genomeDir ${GENOME_DIR} \
        --readFilesIn ${READ1} ${READ2} \
        --readFilesCommand zcat \
        --outFileNamePrefix ${OUTPUT_PREFIX}_ \
        --outSAMtype BAM SortedByCoordinate \
        --outSAMattributes All \
        --outFilterMultimapNmax 1 \
        --outFilterMismatchNmax 3 \
        --alignIntronMax 1 \
        --alignSJDBoverhangMin 1 \
        --alignSJoverhangMin 8 \
        --outFilterScoreMinOverLread 0.66 \
        --outFilterMatchNminOverLread 0.66

   Copy
2. 2

   Run SAILOR to identify editing sites.

   SAILOR vlatest (Inferred with models/gemini-2.5-flash) GitHub

   $ Bash example

   # It's recommended to install SAILOR in a conda environment
   # conda create -n sailor python=3.8
   # conda activate sailor
   # pip install git+https://github.com/yeolab/SAILOR.git
   
   # Define input and output paths
   INPUT_BAM="path/to/your/aligned_reads.bam" # Replace with your input BAM file
   REFERENCE_GENOME="path/to/your/GRCh38.fa" # Placeholder: Path to the reference genome FASTA (e.g., GRCh38)
   KNOWN_SNPS_VCF="path/to/your/common_snps_GRCh38.vcf.gz" # Placeholder: Path to a VCF of known common SNPs (e.g., from dbSNP) to filter out genomic variants
   OUTPUT_PREFIX="sailor_editing_sites"
   
   # Create output directory if it doesn't exist
   mkdir -p sailor_output
   
   # Run SAILOR to identify editing sites
   # Adjust parameters like --min-coverage, --min-edit-fraction, --threads as needed
   SAILOR run \
       --bam "${INPUT_BAM}" \
       --genome "${REFERENCE_GENOME}" \
       --vcf "${KNOWN_SNPS_VCF}" \
       --output "sailor_output/${OUTPUT_PREFIX}" \
       --min-coverage 10 \
       --min-base-quality 20 \
       --min-map-quality 20 \
       --min-edit-fraction 0.1 \
       --threads 8
   

   Copy View on GitHub
3. 3

   Code available at: https://github.com/YeoLab/FLARE

   FLARE vNot specified (Inferred with models/gemini-2.5-flash)

   $ Bash example

   # Install FLARE using conda (uncomment to run)
   # conda create -n flare_env python=3.8
   # conda activate flare_env
   # conda install -c bioconda flare
   
   # Define reference files (example using GRCh38/hg38 and Gencode v38)
   # Replace with actual paths to your reference files
   GENOME_FASTA="path/to/GRCh38.primary_assembly.genome.fa" # e.g., from Gencode or UCSC
   GTF_ANNOTATION="path/to/gencode.v38.annotation.gtf" # e.g., from Gencode
   
   # Define output directories
   FLARE_INDEX_DIR="flare_index_GRCh38_gencode_v38"
   QUANT_OUTPUT_DIR="flare_quant_output"
   
   # Input BAM file (example: replace with your aligned RNA-seq BAM file)
   INPUT_BAM="path/to/your_aligned_rna_seq.bam"
   
   # Create output directories if they don't exist
   mkdir -p "${FLARE_INDEX_DIR}"
   mkdir -p "${QUANT_OUTPUT_DIR}"
   
   # 1. Build the FLARE index
   # This step needs to be run once for a given genome and annotation.
   echo "Building FLARE index..."
   flare build \
       -g "${GENOME_FASTA}" \
       -a "${GTF_ANNOTATION}" \
       -o "${FLARE_INDEX_DIR}" \
       --threads 8 # Example: use 8 threads
   
   # Check if index build was successful
   if [ $? -ne 0 ]; then
       echo "FLARE index build failed. Exiting."
       exit 1
   fi
   
   # 2. Quantify full-length isoforms
   echo "Quantifying full-length isoforms with FLARE..."
   flare quant \
       -i "${FLARE_INDEX_DIR}" \
       -b "${INPUT_BAM}" \
       -o "${QUANT_OUTPUT_DIR}" \
       --threads 8 # Example: use 8 threads
   
   # Check if quantification was successful
   if [ $? -ne 0 ]; then
       echo "FLARE quantification failed. Exiting."
       exit 1
   fi
   
   echo "FLARE analysis complete."

   Copy
4. 4

   Run FLARE to identify regions of enriched editing.

   FLARE (Inferred with models/gemini-2.5-flash) v0.1.0 (Inferred from setup.py in yeolab/FLARE) GitHub

   $ Bash example

   # Clone the FLARE repository if not already available
   # git clone https://github.com/yeolab/FLARE.git
   # cd FLARE
   
   # Define input and output files
   # INPUT_BAM: Aligned RNA-seq reads in BAM format (e.g., from STAR alignment)
   INPUT_BAM="path/to/your/aligned_reads.bam"
   # OUTPUT_PREFIX: Prefix for all output files generated by FLARE
   OUTPUT_PREFIX="flare_output"
   
   # Define reference datasets
   # GENOME_FASTA: Reference genome in FASTA format (e.g., hg38.fa)
   # Source: UCSC Genome Browser (e.g., http://hgdownload.soe.ucsc.edu/goldenPath/hg38/bigZips/hg38.fa.gz)
   GENOME_FASTA="path/to/reference/hg38.fa"
   # REPEATS_BED: Repeat regions in BED format (e.g., from RepeatMasker)
   # This file is often provided with genome builds or can be generated.
   REPEATS_BED="path/to/reference/repeats.bed"
   # KNOWN_SITES_VCF: Known RNA editing sites in VCF format (optional, but recommended for filtering/annotation)
   # Source: e.g., REDIportal (http://www.www.compgen.unibe.ch/REDIportal/download.html) or DARNED
   KNOWN_SITES_VCF="path/to/reference/known_editing_sites.vcf"
   
   # Run FLARE to identify regions of enriched editing
   # Adjust parameters like --min_coverage, --min_editing_ratio, etc., as needed
   python flare.py \
       -i "${INPUT_BAM}" \
       -o "${OUTPUT_PREFIX}" \
       -g "${GENOME_FASTA}" \
       -r "${REPEATS_BED}" \
       -s "${KNOWN_SITES_VCF}" \
       --min_coverage 10 \
       --min_editing_ratio 0.05 \
       --min_base_quality 20 \
       --min_mapping_quality 20 \
       --threads $(nproc)
   

   Copy View on GitHub
5. 5

   Code available at: https://github.com/YeoLab/FLARE

   FLARE vv0.1.0 GitHub

   $ Bash example

   # Installation (example, uncomment if needed)
   # pip install flare
   
   # Example config.yaml for FLARE
   # This configuration file specifies input BAM files, GTF annotation,
   # and output directories for FLARE analysis.
   # Replace placeholders with actual paths.
   cat << EOF > config.yaml
   # Path to the GTF annotation file
   gtf: /path/to/your/reference/genome.gtf
   
   # Directory containing input BAM files
   bam_dir: /path/to/your/bam_files
   
   # List of sample names (corresponding to BAM files in bam_dir, e.g., sample1.bam)
   samples:
     - sample1
     - sample2
   
   # Output directory for FLARE results
   output_dir: ./flare_output
   
   # Optional: Number of threads to use
   threads: 8
   
   # Optional: Other FLARE specific parameters can be added here
   # For example, minimum read count, minimum junction count, etc.
   # min_read_count: 5
   # min_junction_count: 3
   EOF
   
   # Execute FLARE with the configuration file
   # Ensure 'flare' is in your PATH or specify the full path to the executable.
   flare run --config config.yaml

   Copy View on GitHub

Tools Used